Key points are not available for this paper at this time.
Over recent decades, the number of available clinical practice guidelines has grown enormously. These guidelines are increasingly used in health-care systems throughout the world to improve the quality of patient care, and this also applies to cancer care. Evidencebased guidelines are seen by professionals, authorities, managers and policy makers as powerful tools for achieving effective and efficient care They are considered to be the ideal mediator for bridging the gap between the growing stream of research findings and actual clinical practice. Guidelines should meet specific quality criteria to ensure good quality. Users should be confident that potential biases inherent in guideline development have been addressed appropriately and that the recommendations for practice are both internally and externally valid, as well as feasible for practice (AGREE (Appraisal of Guidelines Research and Evaluation) Collaborative Group, 2000). However, recent studies have reported that the methodological quality of many guidelines is modest and is heterogeneous between the different guidelines and different guideline programmes Although clinical guidelines can provide a solution to some of the important problems in patient care, there are issues that need to be tackled before guidelines can achieve their full potential (Grol, 2001a). We will start by outlining these problems, and then we will present a set of criteria for high-quality guidelines developed and validated by an international group of researchers and guideline developers (the AGREE collaboration). Some cancer guidelines (including those produced by the French National Federation of Cancer Centres-FNCLCC the SOR) were used in the validation process for these criteria. We will then provide some recommendations for guideline developers with the aim that this will help researchers and practitioners in cancer care to develop highquality guidelines for the management of their patients.
Grol et al. (Fri,) studied this question.